trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9911504424778761,
  "eval_steps": 500,
  "global_step": 84,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.5899705014749262,
      "grad_norm": 0.681498646736145,
      "learning_rate": 8.095238095238097e-06,
      "loss": 1.8037,
      "step": 50
    },
    {
      "epoch": 0.9911504424778761,
      "eval_accuracy": 0.5884438079508274,
      "eval_f1": 0.0681875209652327,
      "eval_loss": 1.2872825860977173,
      "eval_precision": 0.2979135893648449,
      "eval_recall": 0.03849973902020729,
      "eval_runtime": 176.3954,
      "eval_samples_per_second": 140.162,
      "eval_steps_per_second": 0.55,
      "step": 84
    },
    {
      "epoch": 0.9911504424778761,
      "step": 84,
      "total_flos": 1.1241305074040832e+16,
      "train_loss": 1.6365551267351424,
      "train_runtime": 2060.6389,
      "train_samples_per_second": 41.993,
      "train_steps_per_second": 0.041
    }
  ],
  "logging_steps": 50,
  "max_steps": 84,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 50,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.1241305074040832e+16,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}